HIV-1 integrase (IN) is an uncommon target for experimental inhibition of HIV-1 infection. We propose to study protection from HIV-1 in vitro and in vivo using recombinant SV40-based vectors (rSV40) to deliver single chain Fv antibodies (SFv) against IN. We have found that rSV40 delivery of the anti- IN SFv, Aw, effectively blocks both IN activity and HIV-1 infection, in vitro in unselected human T cells, and in vivo in SCID-hu mice bearing human thymic implants. With Dr. Harris Goldstein, who has used his SCID-hu mouse model system to study HIV-1 infection, we will test single agent and combination genetic therapy using Aw and a second, complementary anti-IN SFv, SFv number 4. SV(Aw) + SV(IN4) inhibition of HIV-1 infection will be analyzed in vitro and in vivo. This proposal is based on this hypothesis: Combinations of anti-IN SFv delivered by rSV40 vectors to mature and pro-genitor cells will protect susceptible cells from HIV-1 in vitro and in vivo. To test this hypothesis, we propose 5 Specific Aims. We will test anti-HIV-1 protection by both anti- IN SFv, delivered by rSV40 vectors, and compare levels and duration of protection from HIV-1 achieved with these vectors in unselected cells to those achieved with retroviral vectors delivering the same transgenes to selected cells. Additive protection should be possible using SV(Aw) and SV(IN4) in combination, and this will be examined. Single and multiple rSV40-transduction of human hematopoietic progenitor cells and protection of their mature, HIV-1-susceptible progeny will be examined in vivo in SCID-hu mice. Gene delivery to human thymic implants in SCID-hu mice will be used to confirm in vivo protection from HIV-1 infection afforded by this approach and to measure the levels of inhibition of HIV-1 that can be achieved by both rSV40 transduction and HIV-1 challenge in vivo. Effects of transduction on T cell development and physiology will be tested. We will prepare for testing anti-IN genetic therapy of immunosuppressive lentivirus in nonhuman primates by developing and testing in vitro a hybrid SIV-based virus containing HIV-1 IN. Finally, we propose quantitative studies to define the frequency and characteristics of the integration of rSV40 DNA into the host chromosomes. SV40-based gene delivery is highly efficient in transducing HIV-1-susceptible cells and their progenitors. Thus, it has great promise, particularly when combined with intracellular SFv, in inhibiting HIV-1 infection. Studies proposed here will help determine the potential efficacy and implications of this approach to intracellular immunization in the treatment of HIV-1 infection.